Related Experiment Video
Updated: Jun 30, 2026
08:36
Empirical, Metagenomic, and Computational Techniques Illuminate the Mechanisms by which Fungicides Compromise Bee Health
Published on: October 9, 2017
10.2K
Run-length compressed metagenomic read classification with SMEM-finding and tagging.
Lore Depuydt1, Omar Y Ahmed2, Jan Fostier1
1Department of Information Technology - IDLab, Ghent University - imec, 9052 Gent, Belgium.
Iscience
|January 7, 2026
Summary
We developed a new computational method for metagenomic read classification using a compressed BWT index. This approach efficiently classifies sequencing data, improving accuracy and speed over existing tools.
Area of Science:
- Computational biology
- Bioinformatics
- Genomics
Background:
- Metagenomic read classification is crucial for analyzing complex biological samples.
- Current methods face challenges with large-scale, diverse sequencing data.
Purpose of the Study:
- To introduce an efficient and accurate method for metagenomic read classification.
- To leverage run-length compression and BWT-based indexing for improved performance.
Main Methods:
- Utilized a run-length compressed Burrows-Wheeler Transform (BWT) index with a move structure.
- Implemented identification of super-maximal exact matches (SMEMs) and a consensus algorithm for classification.
- Employed a sampled tag array to associate SMEMs with class identifiers.
Main Results:
- Achieved superior accuracy and runtime performance compared to SPUMONI 2.
- Demonstrated competitive memory efficiency against Cliffy.
- Successfully classified both long and short reads across diverse datasets.
Conclusions:
- The proposed method offers an efficient and accurate solution for metagenomic read classification.
- Run-length compressed BWT-based indexing with full SMEMs is effective for large-scale data analysis.
